Oceanography and Marine Biology: An Annual Review, 2005, 43, 279-418 © R. N. Gibson, R. J. A. Atkinson, and J. D. M. Gordon, Editors Taylor & Francis THE ECOLOGY OF RAFTING IN THE MARINE ENVIRONMENT. II. THE RAFTING ORGANISMS AND COMMUNITY MARTIN THIEL1,2* & LARS GUTOW3 1Facultad Ciencias del Mar, Universidad Católica del Norte, Larrondo 1281, Coquimbo, Chile 2Centro de Estudios Avanzados en Zonas Áridas (CEAZA), Coquimbo, Chile *E-mail: [email protected]; Fax: ++ 56 51 209 812 3Alfred Wegener Institute for Polar and Marine Research, Biologische Anstalt Helgoland, Box 180, 27483 Helgoland, Germany E-mail: [email protected] * author for correspondence Abstract Rafting of marine and terrestrial organisms has been reported from a variety of substrata and from all major oceans of the world. Herein we present information on common rafting organisms and on ecological interactions during rafting voyages. An extensive literature review revealed a total of 1205 species, for which rafting was confirmed or inferred based on distributional or genetic evidence. Rafting organisms comprised cyanobacteria, algae, protists, invertebrates from most marine but also terrestrial phyla, and even a few terrestrial vertebrates. Marine hydrozoans, bryozoans, crustaceans and gastropods were the most common taxa that had been observed rafting. All major feeding types were represented among rafters, being dominated by grazing/boring and suspension-feeding organisms, which occurred on all floating substrata. Besides these principal trophic groups, predators/scavengers and detritus feeders were also reported. Motility of rafting organisms was highest on macroalgae and lowest on abiotic substrata such as plastics and volcanic pumice. Important trends were revealed for the reproductive biology of rafting organisms. A high proportion of clonal organisms (Cnidaria and Bryozoa) featured asexual reproduction, often in combination with sexual reproduction. Almost all rafting organisms have internal fertilisation, which may be due to the fact that gamete concentrations in the rafting environment are too low for successful fertilisation of external fertilisers. Following fertilisation, many rafting organisms incu- bate their offspring in/on their body or deposit embryos in egg masses on rafts. Local recruitment, where offspring settle in the immediate vicinity of parents, is considered an important advantage for establishing persistent local populations on a raft, or in new habitats. Some organisms are obligate rafters, spending their entire life cycle on a raft, but the large majority of reported rafters are considered facultative rafters. These organisms typically live in benthic (or terrestrial) habitats, but may become dispersed while being confined to a floating item. Substratum characteristics (complexity, surface, size) have important effects on the composition of the rafting community. While at sea, ecological interactions (facilitation, competition, predation) contribute to the com- munity succession on rafts. Organisms capable to compete for and exploit resources on a raft (space and food) will be able to persist throughout community succession. The duration of rafting voyages is closely related to rafting distances, which may cover various geographical scales. In chronological order, three features of an organism gain in importance during rafting, these being ability to (1) hold onto floating items, (2) establish and compete successfully and (3) develop persistent local 279 MARTIN THIEL & LARS GUTOW populations during a long voyage. Small organisms that do not feed on their floating substratum and, with asexual reproduction or direct development, combine all these features appear to be most suited for long-distance dispersal on rafts and successful colonisation after reaching new habitats. All available evidence suggests that rafting is an important process for the population dynamics of many organisms and that it also has had and continues to have a strong influence on coastal biodiversity. Introduction … the agency of the Kafirkuils River in bringing down drift of the Riversdale coast at Still Bay after the great flood of November 1928 may be cited. Not only fruits, seeds, and other smaller vegetal matter were so transported, but railway construction plant, boats, ostriches, sheep, burrowing snakes (Typhlops), puff-adders and other Ophidia, lizards, tree-mice, scorpions, species of Coleoptera, and on previous occasions even a baboon or two. Muir (1937) reporting on drift material from beaches of South Africa Anecdotal reports such as this one by Muir (1937) or those of other authors (e.g., Guppy 1917, King 1962, Carlquist 1965, Van Duzer 2004) provide testimony of the diverse kinds of floating items that reach the oceans. Further, these reports give hints that many different organisms utilise these items as floating devices in order to escape from drowning. These observations of travellers on floating items also provoke many questions, the two most relevant ones being: where do these organisms come from, and where might they go? While these questions are simple and straight- forward, finding the answers is not. In fact, biologists in the past have struggled to infer possible answers and in most cases evidence has remained circumstantial. Organisms travelling on floating items over the sea surface may be transported to areas which they might not have reached otherwise. This process, termed rafting, can have important conse- quences if travellers, upon reaching new habitats, are capable of establishing new populations. Rafting is of particular importance for those organisms that are not capable of autonomous dispersal in or across the ocean. This is, for example, the case of many coastal organisms without pelagic dispersal stages. Also most terrestrial organisms are unable to travel over the sea without the aid of a transport vehicle. In spite of a limited capacity for autonomous dispersal, many organisms have wide geographic distributions, even across wide oceanic barriers. This has led biologists to infer that disjunct distribution patterns could be the result of rafting on floating items. Distributional evidence has been reported for a wide diversity of organisms ranging from hydrozoans (Cornelius 1992a), small polychaetes (Knight-Jones & Knight-Jones 1984), molluscs (Ó Foighil et al. 1999, Castilla & Guiñez 2000), echinoderms (Mortensen 1933, Fell 1962 cited in Fell 1967) and crustaceans (Svavarsson 1982, Peck 1994), to terrestrial insects (Abe 1984, Niedbala 1998), reptiles (Raxworthy et al. 2002) and mammals (Hafner et al. 2001). In some of these cases, evidence for rafting as possible dispersal mechanism is better than in others. For example, in the case of the polychaete or mollusc species for which rafting was inferred, these live on/in macroalgae with a high floating potential. However, in other cases, the respective species with disjunct distributions have never been observed rafting and this process has only been suggested due to a lack of alternative explanations. Many organisms have also been observed in flagrante, i.e., while rafting on the high seas. In particular when coastal or terrestrial organisms were found on floating items far from the next shore, authors have suggested that these could potentially travel far distances on their rafts. Polychaetes (Arnaud et al. 1976, Averincev 1980), molluscs (Helmuth et al. 1994), echinoderms 280 RAFTING OF BENTHIC MARINE ORGANISMS (Hobday 2000a) and crustaceans (Ólafsson et al. 2001, Gutow & Franke 2003) are frequently reported from floating items in the open ocean, often at far distances from the nearest coast. Not all organisms are equally adapted to rafting. One of the most important preconditions is that a rafter needs to hold on to the substratum, and not all organisms can cling efficiently to floating items. Edgar & Burton (2000) reported that many epifaunal organisms were rapidly lost from floating macroalgae. Feeding conditions on a floating item may also be very different from those in the natural habitats of most facultative rafters. Some organisms may even be capable of enduring a long journey on a floating item. During such a journey rafting organisms are also exposed to a variety of interactions with their substratum, fellow rafters and the water body in which they are travelling. Floating items provide attachment substratum and in many cases also food for rafters. In a rafting community one can expect similar ecological interactions as in benthic habitats (competition, facilitation, predation) and these may have profound effects on species succession during a long journey. In addition, oceanic conditions will affect rafters in various ways, e.g., in form of abiotic (wave action, temperature and salinity) and biotic (nutrients, food, predators) factors. Floating items travel at the sea surface, where environmental conditions might be considered as extreme for many species. At the water surface both inorganic and organic chemical compounds accumulate and also solar radiation is substantially higher than in the water column immediately below (Zaitsev 1970, Cheng 1975). Organisms on a raft are probably exposed to greater temperature changes and turbulence than encountered in their benthic habitat to which they are adapted (Holm- quist 1994). All these interactions will have an influence on the survival of rafters during their journey and some space will be dedicated to these interactions in our review. Rafting will only be a significant ecological and evolutionary process if rafters are capable of establishing